Clinical outcomes of left atrial appendage occlusion in patients with previous intracranial or gastrointestinal bleeding: Insights from the LOGIC (Left atrial appendage Occlusion in patients with Gastrointestinal or IntraCranial bleeding) International Multicenter Registry

Aims: To compare outcomes of patients who underwent left atrial appendage occlusion (LAAO) for nonvalvular atrial fibrillation (NVAF) and contraindication to anticoagulants due to history of either gastrointestinal (GI) or intracranial (IC) bleeding. Methods: Patients with NVAF that underwent LAAO for GI or IC bleeding from seven centers were included in this observational study. Baseline characteristics, procedural features, and follow-up data were collected, and compared between the two groups. The primary outcomes were incidence of ischemic and hemorrhagic events at 12-month. Results: Six hundred twenty-eight patients were included, 57% with previous GI-bleeding, and 43% with previous IC-bleeding. Median CHA 2 DS 2-VASc score was 4 (interquartile range [IQRs]: 3-5) for both GI-bleeding and IC-bleeding patients, while GI-bleeding patients had a higher HAS-BLED score (4 [IQRs: 3-4] vs. 3 [IQRs]: 2-3]; p = 0.001). At 12-month follow-up, relative risk reduction for stroke was similar between the two groups. The GI-bleeding group had more hemorrhagic events compared to IC-bleeding group (any bleeding 8.4% vs. 3.2%; p = 0.012; major bleeding BARC 3-5: 4.3% vs. 1.8; p = 0.010). At multivariate analysis history of GI bleeding was an independent predictor of hemorrhagic events (adjusted HR: 2.39, 95% confidence interval: 1.02-5.63; p = 0.047). Conclusions: Outcomes after LAAO may be affected by the different indications for the procedure. In our study, GI-bleeding and IC-bleeding as indication to LAAO differ in their baseline characteristics. LAAO confirms its efficacy in ischemic risk reduction in both groups, while GI bleeding seems to be an independent predictor of bleeding recurrence at 12 months behind the antithrombotic regimen.